Upload
others
View
2
Download
0
Embed Size (px)
Citation preview
Managed by UT-Battellefor the Department of Energy
This presentation does not contain any proprietary, confidential, or otherwise restricted information
Novel Low-Temperature Proton Transport Membranes
Andrew Payzant
Oak Ridge National Laboratory
June 6, 2008
Project ID# PDP2
2 Managed by UT-Battellefor the Department of Energy
Overview
• Team members– Andrew Payzant– Beth Armstrong– Tim Armstrong– Junhang Dong, U. Cincinnati
• Timeline– Started 02/2004– Stopped 12/2004– Restarted 04/2007
• Budget– FY04: $100k– FY05: $45k– FY06: $30k– FY07: $300k– FY08: $250k
• Barriers to be addressed– Low H2 flux– Poor mechanical properties– Poisoning by CO2 and H2S
3 Managed by UT-Battellefor the Department of Energy
Objective
• The objective is to develop a novel ceramic proton conductor based on La2Mo2O9 (LAMOX) for use as a H2separation membrane.
• The objective is achieved through:– compositional development;– characterization of the electrical properties, chemical stability,
hydrogen flux, and thermo-mechanical properties;– Neutron diffraction analysis of selected materials to better
understand the hydrogen transport properties; &– evaluation of surface exchange catalysts.
• The goal will be to synthesize thin asymmetric membranes (< 25 μm) from candidate materials with and without exchange catalysts for additional flux testing to determine the range of fluxes possible in these materials.
4 Managed by UT-Battellefor the Department of Energy
Review of performance metrics• Cost per sq foot material:
– Unknown but likely closely similar to other ceramic membranes
• Module cost:– Unknown but likely closely similar to other ceramic membranes
• Flux rate: – Very limited permeation data collected to date at NMT and NETL suggests
fluxes similar to Y-doped BaCeO3, but at significantly lower temperatures.
• %H2 recovery: – To be determined
• Hydrogen quality: – As for all ion-transport systems, should be 100% H2
• Operating temperature, pressure:– T<550°C, P yet to be determined (based on mechanical stability)
• Durability:– Stable at temperature in H2O, CO, and CO2. H2S stability to be determined
5 Managed by UT-Battellefor the Department of Energy
Milestones
• Status as of May 1, 2008: ready for Task 2c: hydrogen flux measurement, in collaboration with Prof. Junhang Dong at the University of Cincinnati
6 Managed by UT-Battellefor the Department of Energy
Lanthanum molybdates: a potential alternative to perovskite ceramics
• Doped cerate proton conductors represent a well-established 20 year old technology, and further property improvements are likely to be modest
• La2Mo2O9 was identified as a fast oxygen ion conductor by researchers in France in 1999/2000
• Focus elsewhere is on oxygen ion conduction - efforts to stabilize high-temperature β-phase by anion and cation doping
• Low temperature α-phase first identified as a possible proton conductor at ORNL in 2003/2004– Conductivity measured at ORNL and Rutgers University– H2 permeance measured at New Mexico Tech (NMT) and NETL
T.R. Armstrong, E.A. Payzant, S.A. Speakman, M. Greenblatt, Low Temperature Proton Conducting Oxide, U.S. Patent Application # US 2006/0292416
7 Managed by UT-Battellefor the Department of Energy
Hydrogen flux is the single most important performance issue
• Present literature suggests that hydrogen flux rates of between 5 and 50 mL/min/cm2 will be required for commercial application of ceramic membranes
• In 2004 we made two preliminary measurements (at NMT and NETL) of hydrogen flux in 3mm thick samples of W-doped and Nb-doped La2Mo2O9
• Results were encouraging: H2 flux was confirmed in both tests, but the magnitude was only 5 x 10-5 mL/min/cm2
• So… how to increase flux by 5 to 6 orders of magnitude?
8 Managed by UT-Battellefor the Department of Energy
Strategies for increasing hydrogen flux• Task 1: Reduce sample thickness
– Reduction from mm to µm dimensions will provide a 2 order of magnitude increase in flux based on the Nerst equation
• Task 2: Increase H2 pressure differential– Initial tests were done at a low Nernst potential: near 1 atm, with dilute H2 on the
source side and inert sweep gas on the other– Gain is logarithmic, so perhaps an order of magnitude gain in flux could be realized
• Task 3: Increase proton conductivity– Alter crystal chemistry with dopants to increase mobility of protons within the
structure, and to alter the ratio of proton to electron conductivity– Potential improvements are not predicable at this time
• Task 4: Increase H2 to proton dissociation rate– Use of surface catalysts has proven effective in the cerate systems and could provide
1-2 order of magnitude increase in flux
• Therefore, flux gains of 4-5 orders of magnitude are realistically achievable in this system, which would make it comparable to established dense ceramic or cermet membranes, but with improved stability
9 Managed by UT-Battellefor the Department of Energy
Dense phase-pure ceramics are made from combustion synthesized powder
The XRD patterns contrast the as-prepared powder (gray) and the fully sintered ceramics (blue) - the latter are phase-pure LAMOX.
4.47
4.57
4.67
4.77
4.87
4.97
5.07
5.17
5.27
5.37
875 925 975 1025 1075 1125
Temperature (°C)
83.2
85.1
87.0
88.8
90.7
92.6
94.4
96.3
98.1
100.0
% T
heo
reti
cal D
en
sity
900/2hr
1000/2hr
1050/2hr
1100/2hr
Undoped LAMOX powders sintered to >95% of theoretical density in 2 hours at temperatures above 1000°C.
10 Managed by UT-Battellefor the Department of Energy
• The high-temperature β-phase La2Mo2O9is a fast oxygen-ion conductor while the α-phase is not.– W-doping stabilizes the β-phase at lower
temperatures, and (at least in homogeneous phase-pure samples) suppresses the proton conduction
• In-situ XRD data from at ORNL reveals:– Below 550°C, La2Mo2O9 relaxes into the α-
phase in air, inert, or hydrogen atmosphere– The α-phase is a proton conductor with a
complex oxygen sublattice ordering compared to the β-phase
– Pr-doping dilates the La2Mo2O9 lattice, but retains the α-phase structure, yielding slightly higher proton conductivity
α-phase
β-phase
αα--phase Laphase La22MoMo22OO99 has a more complex has a more complex structure than structure than perovskiteperovskite or or pyrochlorepyrochlore
11 Managed by UT-Battellefor the Department of Energy
The La2Mo2O9 phase transformations are fully reversible
Undoped LAMOX samples reversibly turn black in hydrogen, white in air, with no structural degradation
Because the phase transformation is related to ordering of oxygen vacancies, it should not result in large intergranular stresses and mechanical failure of the ceramic, nor to debonding from a support with reasonably matched thermal expansion
Thermal Expansion
7.14
7.15
7.16
7.17
7.18
7.19
7.2
7.21
7.22
7.23
7.24
0 100 200 300 400 500 600 700 800
Temperature [°C]
12 Managed by UT-Battellefor the Department of Energy
-9
-8
-7
-6
-5
-4
-3
-2
1 1.2 1.4 1.6 1.8 2 2.2
1000/T (1/K)
heating
in air
initial heating
in humidified, dilute
H2 after soaking at
200°
C for 1 hour
cooling in humidified,
dilute H2
subsequent heating and cooling in
humidified, dilute H2
Lanthanum molybdates exhibit proton conductivity and hydrogen flux
ConductivityHydrogen flux
-10
-9.8
-9.6
-9.4
-9.2
-9
-8.8
-8.6
1.0 1.2 1.4 1.6 1.8 2.0 2.2
1/kT (mol/kJ)
200°C300°C500°C 400°C
cooling
heating
13 Managed by UT-Battellefor the Department of Energy
The α-phase structure is essential for the elevated conductivity in H2
• W-doping can be used to stabilize the high-temperature β-phase LAMOX structure, but the conductivity of the stabilized phase is not sensitive to presence of H2.
• Working to understand why the ordered defect structure is a proton conductor, whereas the disordered defect structure is a better oxygen ion conductor.
14 Managed by UT-Battellefor the Department of Energy
Conductivity of 10% Pr-doped LAMOX is similar to undoped LAMOX
• Pr:LAMOX conductivity same as LAMOX, with same increased conductivity in H2 atmosphere compared to air or inert
• No loss in conductivity from doping, but no gain from increased free volume
Sample mounted for 4-point conductivity measurement
15 Managed by UT-Battellefor the Department of Energy
LaLa22MoMo22OO99 is stable in His stable in H22 and COand CO22
• In-situ HTXRD tests at ORNL demonstrate that La2Mo2O9 is stable in H2 and CO2– Tested for 10+ days in H2 at ~500°C– Tested for 2 days in CO2 at 800°C - no decomposition
16 Managed by UT-Battellefor the Department of Energy
Thin supported membranes of Thin supported membranes of LaLa22MoMo22OO99 have been fabricatedhave been fabricated
• W-doped lanthanum molybdate(W:LAMOX) was screen printed onto porous support.
• Although full density was not realized initially, thin (< 20 μm), adherent, and conformal layers were achieved.
SEM micrograph of a W:LAMOX membrane supported on a porous 8mol% Yttria Stabilized Zirconia(8YSZ) substrate.
17 Managed by UT-Battellefor the Department of Energy
Collection of hydrogen permeation Collection of hydrogen permeation data on Ladata on La22MoMo22OO99 is in progressis in progress
• Suitable 1-inch diameter dense ceramic membranes have been synthesized at ORNL, with gold seals to minimize gas leakage
Hydrogen permeation cell at the University of Cincinnati. Leakage tests have been completed as of 05/01/2008 and permeation tests are in progress
18 Managed by UT-Battellefor the Department of Energy
Progress and Future Work
• Task 1: reduce sample thickness– FY2007: phase-pure bulk powders via combustion synthesis– FY2007: thin samples via tape casting, followed by lamination, pressing and firing– FY2007: inks developed for screen printing ultra-thin membranes on porous supports– FY2008: successful thin (<20µm) membrane deposited on YSZ support
• Task 2: increase H2 pressure differential– Porous supports are needed to provide adequate mechanical strength– FY2008 will establish suitable supports to avoid stresses and diffusion
• Task 3: increase proton conductivity– FY2007/08 improvements in ceramic processing have allowed us to synthesize bulk
quantities of powders, and establish correct sintering conditions to make dense single-phase membranes
– B-site doping: FY2007/08 W improves sinterability but stabilizes β-phase; FY2004 Nbfound to generate problematic impurity phases
– A-site doping: FY2007 Ba appears good choice to increase oxygen vacancy concentration, but FY2008 we find little change in conductivity; FY2007/08 Pr appears good choice with larger lattice parameter and more free volume, but no change in conductivity
• Task 4: increase H2 to proton dissociation rate– No significant gains observed to date for humidified versus dry 4%H2– FY2008 will investigate effect of dispersed Ni or Pd catalysts
19 Managed by UT-Battellefor the Department of Energy
Project Summary• Homogeneous powders routinely made by combustion synthesis from stock
solutions• High quality samples synthesized by dry pressing of powders, or by tape
casting slurries made from these powders, followed by lamination and sintering.
• Tungsten was initially added to improve sinterability, but new undopedsamples can be sintered to near theoretical density without tungsten (due to improved processing)
• Phase analysis by XRD, phase stability by in-situ HTXRD• Conductivity analysis by 4-point van der Pauw method in controlled
temperature and gas environment• New undoped (α-phase) samples confirmed previous conductivity results, but
new W-doped (β-phase) material shows no proton conduction (as expected)• New Pr-doped (α-phase) material shows similar proton conduction to undoped
material• New Ba-doped (α-phase) material shows similar proton conduction to undoped
material• Porous supports being investigated - YSZ considered.
20 Managed by UT-Battellefor the Department of Energy
Acknowledgments
• This program presently includes significant technical support from John Henry, Jr. (ORNL), Robbie Peascoe (ORNL), Claire Chisholm (UTK), and Zhong Tang (UC).
• Past team members at ORNL included Scott Speakman, Robert Carneim, and Glen Kirby.